Tuesday
18th November - Nism up
From
John Storey
The
last 24 hours have been particularly eventful. (These missives
are written directly after dinner, when it's impossible to
move for an hour or so until dinner settles.)
Yesterday
evening began with a bang when we fired up the MISM. (Regular
readers will recall that the power supply card in the NISM
had previously been deemed unsuitable for up-firing, on account
of the self-incineration of the high-side switch). Anyway,
we plugged in the MISM and were immediately greeted by the
cheery little red led on the front panel, and the nice bright
green one on the power supply card itself.
It
took a few seconds before I remembered we don't *have* a green
led on the power supply card, a recollection that was reinfiorced
by the clouds of acrid smoke cominng from the vicicnity of
the high-side switch. Things were not looking good for a)
getting anything to work at all b) high-side switches in general
c) Siemens nuclear-bomb-proof high-side switches in particular
c) the reputation of quality German engineering (the only
other German things in the AASTO are the Sonnenscheins).
When
the smoke cleared it became apparent that all that had happened
was that the brown slime had bridged across from the case
of the high-side switch to ground, and was enjoying its last
moments of glory. A quick squirt with general purpose detergent
(unsuitable for food-preparation surfaces), and a bit of a
scrub and everything was right as rain.
The
mism power supply voltages all came up fine, so we tried to
telnet to it but got no response. Then we plugged the mism
power supply board into the nism and fired that up - it looked
happy but we couldn't telnet to it either.
We
measured lots of voltages, peered at the PC104 stack and reminded
each other of how awful it was taking it apart, typed in random
commands to poodle, wiggled connectors, briefly tried to visualise
how beautiful life could be if computers had never been invented,
and packed it in for the night.
This
year we're lodged in the "Elevated Dorm", the blue
building with the satellite dish on top. It is actually very
comfortable - much more room than the Jamesways, *quiet*,
and with the shower and the Head all in the same building.
It's a comfortable temperature (sort of), and even has cupboards
and drawers. The only downside is having to share a room with
Ant...
Also
they got the phone in the AASTO fixed, so now it's a bit safer
working in there. We still haven't got our fire extinguisher
back, despite asking for it at least twice a day since we
arrived. If it does catch fire we can try throwing snow on
it, but 4,000lbs of liquid propane will probably require quite
a bit of snow.
Ev
Paschal was out at the AASTO bright and early this morning,
and got straight to work trying to figure out what had gone
wrong with the TEG. Ev is great. Not only is he a good bloke,
but he had no hesitation in immediately getting out the sponge
and the general purpose detergent and scrubbing down all the
grunge that we'd deliberately not cleaned from TEG so Ev could
look at it. A PhD engineer who doesn't mind getting his hands
dirty is a valuable asset in Antarctica.
Unfortunately
it quickly became clear that the AASTO is totally and utterly
stuffed. The stainless-steel exhaust pipe is corroded through
completely where it goes through the ceiling, and is completely
blocked with green and yellow indescribable lumps of crud.
Ev took photos of it. We could tell things were bad because
Ev kept saying things like "Oh boy". It seems unlikely
the TEG can be fixed this season, but we'll see what ASA can
do. It's amazing that something as inert as freon can be converted
into an unstoppable stainless-steel-munching fibreglass-dissolving
brown-goo-creating circuit-board-illuminating monster chemical
simply by passing it over platinum beads at a few hundred
degrees. Any of the 50lbs of freon that didn't decompose has
by now wandered off to the stratosphere to munch holes in
the ozone layer.
At
about the time we had the AASTO covered in yet another pile
of yellow and white grunge, and were picking up pieces of
what used to be stainless steel and/or fibreglass, in walks
the station manager, station science manager, and health and
safety officer. They stood around for a bit shaking their
heads and saying things like "smells bad" and "might
be mercaptan" and "oh boy", and than left before
anything bad happened to their health. I reminded them about
the fire extinguisher as they left.
We
then pulled the TEG apart, and it's kind of ugly and corroded
inside. Worse, it's full of rockwool insulation, which is
now over everything. Ev says it's not carcinogenic. I believe
it's made out of basalt, but have no idea how. It may involve
soaking rocks in the brown slime they get out of AASTOs.
Now
for the good news. While we were sleeping Michael Ashley crept
out of his bed, telnetted to the "super", and fixed
up the software!! Apparently we'd been using the wrong version.
With
that fixed, instant success! We lit up the nism and immediately
poodle was able to engage it in animated converstation. The
only problem was that all the nism would send back was rows
of little square boxes. To a geek that means only one thing
- baud rate. Being geeks we changed the baud rate of the super
to 19,200 (a figure we chose at random), and sure enough it
sprung into life.
A
quick check out shows: adc bb = -17.8C adc volts = 25.51 adc
amps = 202.6 mA The rotator works, and calibrates. The cooler
works - 96K after a few minutes. Choptest doesn't work. First,
if you ask for 20 seconds it only takes data for two. Secondly,
it gives the numbers you want, even if they're wrong. I think
it's being overly polite. For example, if I set the chop frequency
to something impossible, "choptest" runs and tells
me everything is just fine.
To
attack this problem further we grabbed a Tektronics digital
cro from MAPO (the Phillips having mysteriously re-appeared,
minus knob, back at UNSW), and probed the chopper driver card.
I *hate* oscilloscopes that are more intelligent than I am.
This one is so smart that took me a good half hour just to
figure out that in fact the chopper is working perfectly,
and generating all the right reference signals. The phase-
locked loop works between about 50Hz and 82Hz.
Now
that dinner has settled (home-made bread, spatchcocks, cous-cous,
fresh lettuce, carrot, plus freshly-ground coffee and lemon-raisin
pie), we'll go out and actually see if the detector works.
If so, we'll have scored one out three working instruments
already.
Cheers,
John
 
|
Further Information
|
Contact:
| 
